Looking around the current technology scene the phrase “Just because we could” seems to be remarkably appropriate. For a simply shocking proportion of the more visible and promoted systems emerging right now there is no anticipated actual purpose, but it is hoped that one will appear after these things are in the market. For marketing purposes it’s vital that potential customers, even if fully aware that something is currently useless, fully believe that what they’re buying into will become useful, or a status symbol, or both. This is what makes Google’s remarkably defensive 10 Myths about Google Glass so astonishing, since the Glass seems to be cracking in public perception before it’s even launched.
Ironically, Glass probably will in the long run prove more useful than what Google announced last week: Android “wearables”. That is watches which are basically displays for your smartphone notifications and a battery life that’s quite likely to be hilariously short. More oddly, the purported uses of a wrist mounted companion to your smartphone just aren’t remotely compelling. We already have many fitness oriented watches in the market, and it’s rumoured that Apple are likely to attack that, however, Google’s proposition was suspiciously light on details regarding sensor data. As such it seems, once again, they’ve fallen into the trap of having created something very clever, but have no idea what to do with it. This is a running theme: the talent that sees that new technology has enabled a new compelling product is remarkably rare and undervalued, so many just make in the hope to stumble on to a goldmine.
Google are just easy to pick on about this, and they are to some extent self-aware, attempting to explore in the open. Another group doing so are the “maker” community, a sort of modern DIY group centred around the possibilities of connecting modern computers to hardware. For various reasons back in the 80s interfacing your computer to a random selection of LEDs or motors wasn’t especially difficult, but as computers became more complicated for a good while between the mid 90s and mid 2000s making your PC or Mac play with arbitrary electronics was essentially out of reach for hobbyists. Then came the Arduino which is a platform you can use to connect various sensors or actuators to a computer, or if your needs are simple even run completely standalone. Many variants of the Arduino exist, and it’s proven very successful.
What’s fascinating about the Arduino community is given the flexibility of what you could theoretically achieve it’s remarkably poor at coming up with stuff that is useful. In fairness, there are counter-examples, such 3D printers connected to PCs via Arduinos, but these exceptions are rare compared to the size of the community. One newly announced variant of the Arduino exemplifies this, and that is the Microview.
This is a heavily integrated Arduino, with an OLED display, all in a package not dissimilar to a chip. Given that they’ve achieved their targets I don’t feel too bad pointing out what I don’t like about this product, and that is it suffers from the Arduino Achilles Heel: lack of connectivity. For a normal Arduino board to connect to something like Bluetooth, Ethernet or WiFi, the component requires an additional (often relatively costly) board known as a “shield”, which is bad enough. The Microview, thanks to its size, is not going to be trivially compatible with such things, so other than circuitry you attach directly (such as your sensors or LEDs) or the USB interface of the programming board, your options for connecting it to the outside world are really very limited, which radically reduces the set of possible problems you could attack with it. A slightly more expensive one with Bluetooth Low Energy support included would be a completely different proposition, but such devices are amazingly resistant to appearing, even with the hype around the Internet of Things.
For all that, I’ve certainly got no room to talk. This is one of my Raspberry Pis:
The Raspberry Pi is essentially a low cost, low power Linux computer with half-way decent ways to connect to simple electronic circuits, and in this case that’s a thermometer. Luckily that Raspberry Pi has Ethernet, and as a result I have it setup to broadcast the temperature over the network every ten seconds. Why? Because I can. The fundamental difference here is that by being connected to the network it will benefit from network effects. My home network has a really stupid number of devices attached to it, and now with very little effort they can get the sensor readings, or whatever else I’m broadcasting. For example, I use the same protocol to control an Android app which simply displays the webpage of urls sent over the network enabling a sort of remote controlled picture frame.
The idea is that adding more components broadcasting their information to the network may eventually result in a sort of critical mass, where new applications emerge. Thermometers for different rooms, virtual lightswitches displayed on tablets, broadcasting the status information for my media players. Metcalfe’s law is that the value of a network is proportional to the square of the number of compatible communicating devices connected to the system, and so I view my direction here as merely applying that principle in the hope of reaching something of value sooner.
These things rarely look useful in isolation, but good applications are the result of combining smaller components together in new ways. If those components don’t have scope for being combined together then they are never going to be more useful than they are individually.
Nigel Birkenshaw runs Atomirex.